chore(pipeline): remove implementação antiga de IndexJobsInValkey comentada

Author: hltav

scraper-go/internal/pipeline/pipeline.go

@@ -87,85 +87,6 @@ func Run(ctx context.Context, adapterList []adapters.Adapter, req models.ScrapeR
 	return dedup.DedupeJobs(allJobs)
 }
 
-// func IndexJobsInValkey(ctx context.Context, rdb *redis.Client, jobs []models.Job, keywords []string) {
-// 	if rdb == nil || len(jobs) == 0 {
-// 		return
-// 	}
-
-// 	globalIndexKey := "scraper:jobs:index"
-// 	indexTTL := 72 * time.Hour
-
-// 	// Limpa os índices de keyword anteriores (composto + sub-termos)
-// 	for _, kw := range keywords {
-// 		sanitizedKw := strings.ToLower(strings.TrimSpace(kw))
-// 		if sanitizedKw == "" {
-// 			continue
-// 		}
-
-// 		normalizedForDel := strings.ReplaceAll(sanitizedKw, "/", " ")
-// 		normalizedForDel = strings.Join(strings.Fields(normalizedForDel), " ")
-// 		rdb.Del(ctx, fmt.Sprintf("scraper:jobs:keyword:%s", normalizedForDel))
-
-// 		for _, term := range strings.Fields(normalizedForDel) {
-// 			rdb.Del(ctx, fmt.Sprintf("scraper:jobs:keyword:%s", term))
-// 		}
-// 	}
-
-// 	for _, job := range jobs {
-// 		// ✅ Usa o ID estável como membro de todos os Sets,
-// 		// igual ao que jobstore.SaveBatch faz — nunca a URL bruta
-// 		id := jobstore.StableID(&job)
-// 		if id == "" {
-// 			continue
-// 		}
-
-// 		// Índice global: apenas IDs (sem duplicar com URLs)
-// 		rdb.SAdd(ctx, globalIndexKey, id)
-
-// 		titleLower := strings.ToLower(job.Title)
-// 		descLower := strings.ToLower(job.Description)
-
-// 		for _, kw := range keywords {
-// 			sanitizedKw := strings.ToLower(strings.TrimSpace(kw))
-// 			if sanitizedKw == "" {
-// 				continue
-// 			}
-
-// 			normalizedKw := strings.ReplaceAll(sanitizedKw, "/", " ")
-// 			subTerms := strings.Fields(normalizedKw)
-
-// 			// Keyword composta: indexa só se todos os sub-termos aparecem na vaga
-// 			matchAll := true
-// 			for _, term := range subTerms {
-// 				if !strings.Contains(titleLower, term) && !strings.Contains(descLower, term) {
-// 					matchAll = false
-// 					break
-// 				}
-// 			}
-
-// 			if matchAll {
-// 				fullKey := fmt.Sprintf("scraper:jobs:keyword:%s", strings.Join(subTerms, " "))
-// 				rdb.SAdd(ctx, fullKey, id)
-// 				rdb.Expire(ctx, fullKey, indexTTL)
-// 			}
-
-// 			// Sub-termos individuais: indexa cada um independentemente
-// 			for _, term := range subTerms {
-// 				if term == "" {
-// 					continue
-// 				}
-// 				if strings.Contains(titleLower, term) || strings.Contains(descLower, term) {
-// 					termKey := fmt.Sprintf("scraper:jobs:keyword:%s", term)
-// 					rdb.SAdd(ctx, termKey, id)
-// 					rdb.Expire(ctx, termKey, indexTTL)
-// 				}
-// 			}
-// 		}
-// 	}
-
-//		rdb.Expire(ctx, globalIndexKey, indexTTL)
-//		slog.Info("Valkey invertido atualizado", "total_vagas", len(jobs))
-//	}
 func IndexJobsInValkey(ctx context.Context, rdb *redis.Client, jobs []models.Job, keywords []string) {
 	if rdb == nil || len(jobs) == 0 {
 		return